No abstract

I. INTRODUCTIONEMTS based on the GaN/AlGaN materials system are rapidly becoming the semiconductor device of choice for RF and power switching applications. These devices require a semi-insulating buffer to suppress leakage and punch-through. RF devices frequently make use of iron (Fe) doping to render the GaN insulating, but for the higher voltages required for many power switching applications, it has been found that carbon (C) doping delivers higher breakdown voltage and lower off-state leakage [1,2]. Unfortunately it has also been found that using carbon can result in a transitory increase in R ON , also known as current-collapse (CC), when switched from the off to the on-state [2,3]. With field plates now universally used to control surface effects, it is clear that the remaining CC in these devices mostly results from charge storage in deep levels in the buffer. Our previous studies have shown that the difference in CC between Fe and C doping results from their acceptor trap levels pinning the bulk Fermi level in the upper and lower halves of the bandgap respectively [4]. GaN:C is p-type with its low hole density, and hence high resistivity, giving long time constants for charging processes (a hole density of only 10 4 cm -3 was inferred in [5]

We report on a floating buffer model to explain "kink," a hysteresis in the output characteristics of Fe-doped AlGaN/GaN HEMTs observed at low drain bias. Unintentionally doped background carbon can make the GaN buffer p-type allowing it to electrically float. We further note that reverse bias trap-assisted leakage across the junction between the 2DEG and the p-type buffer can provide a mechanism for hole injection and buffer discharging at just a few volts above the knee, explaining the "kink" bias dependence and hysteresis. We show that HEMTs with a different background carbon have dramatically different kink behaviors consistent with the model. Positive and negative magnitude drain current transient signals with 0.9-eV activation energy are seen, corresponding to changes in the occupation of carbon acceptors located in different regions of the GaN buffer. The observation of such signals from a single trap calls into question conventional interpretations of these transients based on the bulk 1-D deep-level transient spectroscopy (DLTS) models for GaN devices with floating regions.

The role of buffer traps (identified as C N acceptors through current DLTS) in the off-state leakage and dynamic Ron of 650V rated GaN-on-Si power devices is investigated. The dynamic Ron is strongly voltage-dependent, due to the interplay between the dynamic properties of the C N traps and the presence of space-charge limited current components. This results in a complete suppression of dyn Ron degradation under HTRB conditions between 420V and 850V.

No abstract

Low temperature atomic layer deposition was used to deposit α-Ga2O3 films, which were subsequently annealed at various temperatures and atmospheres. The α-Ga2O3 phase is stable up to 400 o C, which is also the temperature that yields the most intense and sharpest reflection by X-ray diffraction. Upon annealing at 450 o C and above, the material gradually turns into the more thermodynamically stable ε or β phase. The suitability of the materials for solar-blind photodetector applications has been demonstrated with the best responsivity achieved being 1.2 A/W under 240 nm illumination and 10 V bias, for the sample annealed at 400 o C in argon. It is worth noting however that the device performance strongly depends on the annealing conditions, with the device annealed in forming gas behaving poorly. Given that the tested devices have similar microstructure, the discrepancies in device performance are attributed to hydrogen impurities.

Comparison between pulsed and CW large signal RF performance of field-plated β-Ga 2 O 3 MOSFETs has been reported. Reduced self-heating when pulse resulted in a power added efficiency of 12%, drain efficiency of 22.4%, output power density of 0.13 W/mm, and maximum gain up to 4.8 dB at 1 GHz for a 2-μm gate length device. Increased power dissipation for higher V DS and I DS resulted in a degradation in performance, which, thermal simulation showed, could be entirely explained by self-heating. Buffer and surface trapping contributions have been evaluated using gate and drain lag measurements, showing minimal impact on device performance. These results suggest that β-Ga 2 O 3 is a good candidate for future RF applications. Index Terms-Ga 2 O 3 MOSFET, large signal RF, pulsed RF, power added efficiency (PAE), pulsed IV. I. INTRODUCTIONT HE material of β-Ga 2 O 3 with a bandgap of 4.9 eV and large electric breakdown strength of 8 MVcm −1 has garnered great interest in the power conversion community; however, there are also opportunities for RF applications [1], [2]. Ga 2 O 3 features a Baliga's figure of merit (BFOM), which is based on the mobility and bandgap, more than 10× higher than for SiC and 4× higher than for GaN [3]. High breakdown voltages up to 755 V with a high drain current on/off ratio of 10 9 have been demonstrated for lateral Ga 2 O 3 transistors [4]. Johnson's figure of merit (saturation velocity times critical electric field product, vsat•Ec) for high frequency devices is very much comparable to GaN [1], [5].

Dispersion in capacitance and conductance measurements in AlGaN/GaN high-electron mobility transistors is typically interpreted as resulting from interface states. Measurements on varying gate-length devices and a model of an interface-trap-free device are used to demonstrate that the distributed-resistance-induced dispersion is significant for 1-MHz measurements if the gate length exceeds ∼10 µm. Hence, interface state density measurements using the conductance technique need to use shorter gate-length devices in order to avoid this artefact.